Search NASA⌕ Search

SEARCH · Search NASA

Results for “Gradient descent”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 199 records · Page 11

Scalable and Energy-Efficient Methods for Interactive Exploration of Scientific Data

The main scientific contributions of this project are the following novel concepts for multidimensional arrays: shape-based similarity join (SIGMOD 2016), incremental view maintenance (SIGMOD 2017), user-defined stencil functions (HPDC 2017), and distributed caching for in-situ processing (SSDBM 2018). Building on our collaboration with the astrophysics group at LBNL, we applied these techniques to the data generated in the Palomar Transient Factory (PTF) astronomical survey. They played a pivotal role in the first-ever observation of a neutron star merger, which produces gravitational waves and turns out to be the origin of heavy elements, including gold. This has lead to a Science magazine article that has received extensive media coverage on ACM TechNews, Slashdot, FiveThirtyEight, and Quanta Magazine, among others. Additionally, two other articles detailing related aspects of the same discovery have been published in the Astrophysical Journal Letters journal. These publications have more than 3,000 citations according to Google Scholar (as of February 2022). This cross-disciplinary collaboration provided very good opportunities to apply database techniques to real-life scientific problems. The fact that they facilitated major discoveries in astrophysics proves the importance of our research. In addition to the work on multidimensional array databases, this project has also developed stochastic gradient descent (SGD) optimization algorithms for training large scale machine learning models, methods for querying in-situ data, and a database query optimizer based on sketch synopses.

79 ASTRONOMY AND ASTROPHYSICS↗

Scalable Second Order Optimization for Machine Learning

Many machine learning (ML) training tasks are essentially optimization processes that would at first glance appear eminently parallelizable and scalable. However, effective acceleration of these tasks with scalable parallel hardware has proven to be elusive. While standard methods for machine learning, e.g., stochastic gradient descent (SGD) for DNNs, tend to be resource efficient, they appear to be fundamentally sequential in nature.

97 MATHEMATICS AND COMPUTING↗

Reliable and Efficient Machine Learning (Final Technical Report)

Modern scientific experiments generate massive amounts of data at a pace much faster than humans can manually analyze. While machine learning has revolutionized commercial data analysis (such as recommending movies or recognizing faces), applying these tools to complex scientific discovery is challenging because scientific answers must be precise, interpretable, and adhere to physical laws. The research under this project aims to develop new mathematical tools and computer algorithms specifically designed for scientific applications. Major progress has been made in automatically cleaning and deconstructing messy experimental data, analyzing the visual information of physical phenomena, determining the underlying physical variables, and providing rig orous mathematical analysis of interesting algorithms and concepts widely used in machine learning. This project addressed the critical gap between our ability to generate massive scientific data and our ability to extract interpretable information from it. We established mathematical foundations for Scientific Machine Learning (SciML) aimed at effective data analytics and automated discovery. Our work focused on three core objectives: (1) developing reliable feature extraction methods for dynamic high-dimensional data, (2) establishing mathematical foundations for discovering dynamics via neural networks, and (3) creating rigorous optimization techniques for these models. Key outcomes come from two fronts. On the practical side, they include the development of algorithms that significantly enhance the extraction of signals from field data, as well as the capability to handle situations that exhibit smooth variations or physical stretching due to temperature changes. They also include the creation of an automated framework for discovering fundamental state variables from raw experimental data, demonstrating the ability to identify intrinsic physical dimensions without prior knowledge of the governing laws. On the theoretical front, the research results in theoretical advances in Optimal Transport, a widely used notion in SciML, specifically regarding functions with fixed-size nodal sets, provide sharp bounds relevant to uncertainty quantification. Meanwhile, the outcomes also include the establishment of convergence theories for nonlocal gradient descent methods, enabling robust optimization with noisy data in high-dimensional settings commonly encountered in scientific modeling. The project also helps creating opportunities to train the next generation of researchers, equipping them with the necessary technical skills for today’s workplace and preparing them for future advances.

97 MATHEMATICS AND COMPUTING↗

CRCNS22 Learning Rules in the Hippocampus and their Mapping to Neuromorphic Systems (Final Technical Report)

Large scale biologically-realistic computational models are key to investigating the interplay between structure and function in nervous systems, thus paving the way to new clinical methods and neuro-inspired computing solutions. This project focuses on the hippocampus, in particular the CA3-CA1 regions, due to their role in associative learning and memory, pattern separation and completion, and spatial navigation. Investigations into the neuronal organization and learning rule(s) of this circuit can shed light into how declarative memories are formed, stored, recalled and forgotten and inform computational, experimental and clinical neuroscience work. Our project aims at developing a novel data-driven methodology supported by a broad heterogeneous base of neuroscience experimental knowledge and inspired from advances in computer science and engineering. Specifically, this work will benchmark existing and new learning rules within a full-scale spiking neural network simulation of the CA3-CA1 region. The model will be based on an open-source repository, called the Hippocampome, which contains neuronal morphologies, firing patterns, synapse probabilities, and most other required parameters for all known neuron types in the rodent hippocampal formation. The model will be first trained in a supervised fashion for associative memory tasks using backpropagation through time traditionally used in computer science, enhanced with a new technique called the surrogate gradient method. This optimization method will be used to obtain a global loss minimization, but it is not biologically inspired as it assumes the use of data not locally available to the synapses. However, we propose its use as a benchmarking tool, to compare the training performance of local biologically plausible and hardware-mappable learning rules at scale. New rules or combinations will be proposed and tested as needed, based on the obtained results. Progress in this area will also drive the development of novel hardware-mappable algorithms for continual lifelong learning and categorization of new events from few presented examples. This project goes beyond the existing state-of-the-art by looking at large scale realistic neuronal circuits as networks trainable via global optimization methods such as surrogate gradient descent. The objective function of the brain that supports learning is largely unknown, but it is likely that it operates through local learning rules. Studying network trajectories around local minima as proposed in this work represents a useful strategy for understanding whether a network is training by using a specific (set of) learning rule(s). Starting from a completely untrained network is a challenging test since it is difficult to determine how the learning rule affects the trajectory of the network. This interdisciplinary project will help understand what rule governs learning in these regions or if multiple learning rules are involved. The work will develop a robust methodology to measure if the network is converging to the target solution, oscillating around it, or diverging away.

59 BASIC BIOLOGICAL SCIENCES↗

A Testbed of Parallel Kernels for Computer Science Research

For several decades, computer scientists have sought guidance on how to evolve architectures, languages, and programming models for optimal performance, efficiency, and productivity. Unfortunately, this guidance is most often taken from the existing software/hardware ecosystem. Architects attempt to provide micro-architectural solutions to improve performance on fixed binaries. Researchers tweak compilers to improve code generation for existing architectures and implementations, and they may invent new programming models for fixed processor and memory architectures and computational algorithms. In today's rapidly evolving world of on-chip parallelism, these isolated and iterative improvements to performance may miss superior solutions in the same way gradient descent optimization techniques may get stuck in local minima. In an initial study, we have developed an alternate approach that, rather than starting with an existing hardware/software solution laced with hidden assumptions, defines the computational problems of interest and invites architects, researchers and programmers to implement novel hardware/ software co-designed solutions. Our work builds on the previous ideas of computational dwarfs, motifs, and parallel patterns by selecting a representative set of essential problems for which we provide: An algorithmic description; scalable problem definition; illustrative reference implementations; verification schemes. For simplicity, we focus initially on the computational problems of interest to the scientific computing community but proclaim the methodology (and perhaps a subset of the problems) as applicable to other communities. We intend to broaden the coverage of this problem space through stronger community involvement. Previous work has established a broad categorization of numerical methods of interest to the scientific computing, in the spirit of the NAS Benchmarks, which pioneered the basic idea of a 'pencil and paper benchmark' in the 1990s. The initial result of the more modern study was the seven dwarfs, which was subsequently extended to 13 motifs. These motifs have already been useful in defining classes of applications for architecture-software studies. However, these broad-brush problem statements often miss the nuance seen in individual kernels. For example, the computational requirements of particle methods vary greatly between the naive (but more accurate) direct calculations and the particle-mesh and particle-tree codes. Thus we commenced our study with an enumeration of problems, but then proceeded by providing not only reference implementations for each problem, but more importantly a mathematical definition that allows one to escape iterative approaches to software/hardware optimization. To ensure long term value, we have augmented each of our reference implementations with both a scalable problem generator and a verification scheme. In a paper we have prepared that documents our efforts, we describe in detail this process of problem definition, scalable input creation, verification, and implementation of reference codes for the scientific computing domain. Table 1 enumerates and describes the level of support we've developed for each kernel. We group these important kernels using the Berkeley dwarfs/motifs taxonomy using a red box in the appropriate column. As kernels become progressively complex, they build upon other, simpler computational methods. We note this dependency via orange boxes. After enumeration of the important numerical problems, we created a domain-appropriate high-level definition of each problem. To ensure future endeavors are not tainted by existing implementations, we specified the problem definition to be independent of both computer architecture and existing programming languages, models, and data types. Then, to provide context as to how such kernels productively map to existing architectures, languages and programming models, we produced reference implementations for most of the kernel (see table). These sample codes should be viewed as 'hints,' designed to show how other designers have mapped a problem's operands and operators to existing hardware and software. Since we wanted such implementations to be illustrative, we tried to ensure they were the most straightforward implementation in the easiest to understand languages using familiar architectures. To that end, most of the linear algebra-oriented computations are written in MATLAB using array indexing to process matrices, rather than one-line library calls to compute the same kernel. This ensures that the kernel's computation is explicit and readable in the implementation and not hidden behind a library. For other problems, such as the Barnes-Hut n-Body solver, the implementations were written in pure C, without any supporting library computations. Along this line, we have also created a scalable problem generator to accompany each computation.

97 MATHEMATICS AND COMPUTING↗

An Adaptive Optimizer for Measurement-Frugal Variational Algorithms

Variational hybrid quantum-classical algorithms (VHQCAs) have the potential to be useful in the era of near-term quantum computing. However, recently there has been concern regarding the number of measurements needed for convergence of VHQCAs. Here, we address this concern by investigating the classical optimizer in VHQCAs. We introduce a novel optimizer called individual Coupled Adaptive Number of Shots (iCANS). This adaptive optimizer frugally selects the number of measurements (i.e., number of shots) both for a given iteration and for a given partial derivative in a stochastic gradient descent. We numerically simulate the performance of iCANS for the variational quantum eigensolver and for variational quantum compiling, with and without noise. In all cases, and especially in the noisy case, iCANS tends to out-perform state-of-the-art optimizers for VHQCAs. We therefore believe this adaptive optimizer will be useful for realistic VHQCA implementations, where the number of measurements is limited.

97 MATHEMATICS AND COMPUTING↗

General-Purpose Bayesian Tensor Learning With Automatic Rank Determination and Uncertainty Quantification

A major challenge in many machine learning tasks is that the model expressive power depends on model size. Low-rank tensor methods are an efficient tool for handling the curse of dimensionality in many large-scale machine learning models. The major challenges in training a tensor learning model include how to process the high-volume data, how to determine the tensor rank automatically, and how to estimate the uncertainty of the results. While existing tensor learning focuses on a specific task, this paper proposes a generic Bayesian framework that can be employed to solve a broad class of tensor learning problems such as tensor completion, tensor regression, and tensorized neural networks. We develop a low-rank tensor prior for automatic rank determination in nonlinear problems. Our method is implemented with both stochastic gradient Hamiltonian Monte Carlo (SGHMC) and Stein Variational Gradient Descent (SVGD). We compare the automatic rank determination and uncertainty quantification of these two solvers. We demonstrate that our proposed method can determine the tensor rank automatically and can quantify the uncertainty of the obtained results. We validate our framework on tensor completion tasks and tensorized neural network training tasks.

Bayesian inference↗

Multiresolution GPC-Structured Control of a Single-Loop Cold-Flow Chemical Looping Testbed

Chemical looping is a near-zero emission process for generating power from coal. It is based on a multi-phase gas-solid flow and has extremely challenging nonlinear, multi-scale dynamics with jumps, producing large dynamic model uncertainty, which renders traditional robust control techniques, such as linear parameter varying H ∞ design, largely inapplicable. This process complexity is addressed in the present work through the temporal and the spatiotemporal multiresolution modeling along with the corresponding model-based control laws. Namely, the nonlinear autoregressive with exogenous input model structure, nonlinear in the wavelet basis, but linear in parameters, is used to identify the dominant temporal chemical looping process dynamics. The control inputs and the wavelet model parameters are calculated by optimizing a quadratic cost function using a gradient descent method. The respective identification and tracking error convergence of the proposed self-tuning identification and control schemes, the latter using the unconstrained generalized predictive control structure, is separately ascertained through the Lyapunov stability theorem. The rate constraint on the control signal in the temporal control law is then imposed and the control topology is augmented by an additional control loop with self-tuning deadbeat controller which uses the spatiotemporal wavelet riser dynamics representation. The novelty of this work is three-fold: (1) developing the self-tuning controller design methodology that consists in embedding the real-time tunable temporal highly nonlinear, but linearly parametrizable, multiresolution system representations into the classical rate-constrained generalized predictive quadratic optimal control structure, (2) augmenting the temporal multiresolution loop by a more complex spatiotemporal multiresolution self-tuning deadbeat control loop, and (3) demonstrating the effectiveness of the proposed methodology in producing fast recursive real-time algorithms for controlling highly uncertain nonlinear multiscale processes. The latter is shown through the data from the implemented temporal and augmented spatiotemporal solutions of a difficult chemical looping cold flow tracking control problem.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

HGQ: High Granularity Quantization for Real-time Neural Networks on FPGAs

Neural networks with sub-microsecond inference latency are required by many critical applications. Targeting such applications deployed on FPGAs, we present High Granularity Quantization (HGQ), a quantization-aware training framework that optimizes parameter bit-widths through gradient descent. Unlike conventional methods, HGQ determines the optimal bit-width for each parameter independently, making it suitable for hardware platforms supporting heterogeneous arbitrary precision arithmetic. In our experiments, HGQ shows superior performance compared to existing network compression methods, achieving orders of magnitude reduction in resource consumption and latency while maintaining the accuracy on several benchmark tasks. These improvements enable the deployment of complex models previously infeasible due to resource or latency constraints. HGQ is open-source and is used for developing next-generation trigger systems at the CERN ATLAS and CMS experiments for particle physics, enabling the use of advanced machine learning models for real-time data selection with sub-microsecond latency.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Maximum likelihood estimation of parameterized 3-D surfaces using a moving camera

A new approach is introduced to estimating object surfaces in three-dimensional space from a sequence of images. A surface of interest here is modeled as a 3-D function known up to the values of a few parameters. The approach will work with any parameterization. However, in work to date researchers have modeled objects as patches of spheres, cylinders, and planes - primitive objects. These primitive surfaces are special cases of 3-D quadric surfaces. Primitive surface estimation is treated as the general problem of maximum likelihood parameter estimation based on two or more functionally related data sets. In the present case, these data sets constitute a sequence of images taken at different locations and orientations. A simple geometric explanation is given for the estimation algorithm. Though various techniques can be used to implement this nonlinear estimation, researches discuss the use of gradient descent. Experiments are run and discussed for the case of a sphere of unknown location. These experiments graphically illustrate the various advantages of using as many images as possible in the estimation and of distributing camera positions from first to last over as large a baseline as possible. Researchers introduce the use of asymptotic Bayesian approximations in order to summarize the useful information in a sequence of images, thereby drastically reducing both the storage and amount of processing required.

Hung, Y.↗

Connectionist model-based stereo vision for telerobotics

Autonomous stereo vision for range measurement could greatly enhance the performance of telerobotic systems. Stereo vision could be a key component for autonomous object recognition and localization, thus enabling the system to perform low-level tasks, and allowing a human operator to perform a supervisory role. The central difficulty in stereo vision is the ambiguity in matching corresponding points in the left and right images. However, if one has a priori knowledge of the characteristics of the objects in the scene, as is often the case in telerobotics, a model-based approach can be taken. Researchers describe how matching ambiguities can be resolved by ensuring that the resulting three-dimensional points are consistent with surface models of the expected objects. A four-layer neural network hierarchy is used in which surface models of increasing complexity are represented in successive layers. These models are represented using a connectionist scheme called parameter networks, in which a parametrized object (for example, a planar patch p=f(h,m sub x, m sub y) is represented by a collection of processing units, each of which corresponds to a distinct combination of parameter values. The activity level of each unit in the parameter network can be thought of as representing the confidence with which the hypothesis represented by that unit is believed. Weights in the network are set so as to implement gradient descent in an energy function.

Hoff, William↗

Simplified Learning Scheme For Analog Neural Network

Synaptic connections adjusted one at a time in small increments. Simplified gradient-descent learning scheme for electronic neural-network processor less efficient than better-known back-propagation scheme, but offers two advantages: easily implemented in circuitry because data-access circuitry separated from learning circuitry; and independence of data-access circuitry makes possible to implement feedforward as well as feedback networks, including those of multiple-attractor type. Important in such applications as recognition of patterns.

Eberhardt, Silvio P.↗

30 years of adaptive neural networks - Perceptron, Madaline, and backpropagation

Fundamental developments in feedforward artificial neural networks from the past thirty years are reviewed. The history, origination, operating characteristics, and basic theory of several supervised neural-network training algorithms (including the perceptron rule, the least-mean-square algorithm, three Madaline rules, and the backpropagation technique) are described. The concept underlying these iterative adaptation algorithms is the minimal disturbance principle, which suggests that during training it is advisable to inject new information into a network in a manner that disturbs stored information to the smallest extent possible. The two principal kinds of online rules that have developed for altering the weights of a network are examined for both single-threshold elements and multielement networks. They are error-correction rules, which alter the weights of a network to correct error in the output response to the present input pattern, and gradient rules, which alter the weights of a network during each pattern presentation by gradient descent with the objective of reducing mean-square error (averaged over all training patterns).

Widrow, Bernard↗

Optical neural network system for pose determination of spinning satellites

An optical neural network architecture and algorithm based on a Hopfield optimization network are presented for multitarget tracking. This tracker utilizes a neuron for every possible target track, and a quadratic energy function of neural activities which is minimized using gradient descent neural evolution. The neural net tracker is demonstrated as part of a system for determining position and orientation (pose) of spinning satellites with respect to a robotic spacecraft. The input to the system is time sequence video from a single camera. Novelty detection and filtering are utilized to locate and segment novel regions from the input images. The neural net multitarget tracker determines the correspondences (or tracks) of the novel regions as a function of time, and hence the paths of object (satellite) parts. The path traced out by a given part or region is approximately elliptical in image space, and the position, shape and orientation of the ellipse are functions of the satellite geometry and its pose. Having a geometric model of the satellite, and the elliptical path of a part in image space, the three-dimensional pose of the satellite is determined. Digital simulation results using this algorithm are presented for various satellite poses and lighting conditions.

Lee, Andrew↗

Inverse problems in diffraction

A two-dimensional problem of diffraction of a plane electromagnetic wave on a smooth 2 pi-periodic surface is considered. A numerical algorithm solving this problem is developed. An inverse problem of determination of the shape of 2 pi-periodic surface using the performance data of reverse scattering is considered. The inverse problem was solved by means of minimization of the residual functional with the help of the gradient descent method. The initial data were calculated with the help of the numerical method. On each step of the iterative method of minimization, the residual functional was calculated approximately with the help of the small slope method. The examples of the shape determination are considered.

Mikheev, Andrew G.↗

Learning and tuning fuzzy logic controllers through reinforcements

A new method for learning and tuning a fuzzy logic controller based on reinforcements from a dynamic system is presented. In particular, our Generalized Approximate Reasoning-based Intelligent Control (GARIC) architecture: (1) learns and tunes a fuzzy logic controller even when only weak reinforcements, such as a binary failure signal, is available; (2) introduces a new conjunction operator in computing the rule strengths of fuzzy control rules; (3) introduces a new localized mean of maximum (LMOM) method in combining the conclusions of several firing control rules; and (4) learns to produce real-valued control actions. Learning is achieved by integrating fuzzy inference into a feedforward network, which can then adaptively improve performance by using gradient descent methods. We extend the AHC algorithm of Barto, Sutton, and Anderson to include the prior control knowledge of human operators. The GARIC architecture is applied to a cart-pole balancing system and has demonstrated significant improvements in terms of the speed of learning and robustness to changes in the dynamic system's parameters over previous schemes for cart-pole balancing.

Berenji, Hamid R.↗

Fast temporal neural learning using teacher forcing

A neural network is trained to output a time dependent target vector defined over a predetermined time interval in response to a time dependent input vector defined over the same time interval by applying corresponding elements of the error vector, or difference between the target vector and the actual neuron output vector, to the inputs of corresponding output neurons of the network as corrective feedback. This feedback decreases the error and quickens the learning process, so that a much smaller number of training cycles are required to complete the learning process. A conventional gradient descent algorithm is employed to update the neural network parameters at the end of the predetermined time interval. The foregoing process is repeated in repetitive cycles until the actual output vector corresponds to the target vector. In the preferred embodiment, as the overall error of the neural network output decreasing during successive training cycles, the portion of the error fed back to the output neurons is decreased accordingly, allowing the network to learn with greater freedom from teacher forcing as the network parameters converge to their optimum values. The invention may also be used to train a neural network with stationary training and target vectors.

Toomarian, Nikzad↗

Dynamic compensatory pattern matching in a fuzzy rule-based control system

A dynamic compensatory matching procedure is suggested as a method to generate an aggregated measure for evaluating the appropriateness of rules for control systems. It is a dynamic weighted matching technique which takes into account incomplete information under real-time requirements. The initial weights of importance of variables are generated with a generalized neural network architecture and a gradient descent algorithm. An intuitive compensatory scheme based on correlations among input variables of training data is adopted so that the system is coherent to a noisy environment.

Sun, Chuen-Tsai↗